基于改进A*算法的移动机器人路径规划  

Path Planning for Mobile Robot Based on Improved A Algorithm

作　　者：汪四新 谭功全 蒋沁 苏畅 WANG Si-xin;TAN Gong-quan;JIANG Qin;SU Chang(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Zigong Sichuan 643000,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川自贡643000

出　　处：《计算机仿真》2021年第9期386-389,404,共5页Computer Simulation

基　　金：人工智能四川省重点实验室开放基金(2019RYJ08)。

摘　　要：针对传统A*算法在移动机器人路径规划中,存在的搜索效率低、路径优化等问题,提出了一种改进的双向A*路径规划算法。改进后的算法采用双向A*寻路搜索,吸取人工势场法引力思想,设定双向搜索中心虚拟目标点,建立从搜索起点到双向搜索的中点的距离衰减函数,吸引双向搜索能在中间区域相遇。并对预估函数加入适当的权值函数,在保障了较优路径前提下加快了算法的收敛速度。考虑机器人实际尺寸问题,进行避障策略设定,在障碍物拐角处进行领域扩展切换,提高了路径安全性。为验证该方法的有效性,通过MATLAB软件进行仿真,结果表明了改进后的方法能更快地搜索出一条无碰撞较优路径。Aiming at the problems of low search efficiency and path optimization in traditional a * algorithm in mobile robot path planning, an improved two-way a * path planning algorithm is proposed.The improved algorithm adopts two-way A* path search, absorbs the gravity idea of artificial potential field method, sets the virtual target point of two-way search center, and establishes the distance attenuation function from the search starting point to the midpoint of two-way search, so as to attract two-way search to meet in the middle area.The appropriate weight function is added to the estimation function, which accelerates the convergence speed of the algorithm under the premise of ensuring a better path.The obstacle avoidance strategy is set based on the actual robot size.Field expansion switching at obstacle corners improves path safety.In order to verify the effectiveness of the method, the simulation experiments were carried out by MATLAB software.The results show that the improved method can search for a collision-free optimal path faster.

关 键 词：算法 路径规划 双向搜索 虚拟牵引点 避障策略 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]